Perception of voice gender in high functioning autism
Posted Oct 03 2008 11:31am
A review of: Groen, W.B., Orsouw, L., Zwiers, M., Swinkels, S., Gaag, R.J., Buitelaar, J.K. (2008). Gender in Voice Perception in Autism. Journal of Autism and Developmental Disorders DOI: 10.1007/s10803-008-0572-8
Social perception using non-verbal means, such as recognition of the social information embedded into facial expressions, voices, and gestures, play a key role in social communication. Unlike typically developing kids, babies that are later diagnosed with autism do not show a preference for their mother’s voice as apposed to other voice sound. The authors of this study reviewed this and other evidence of atypical sound processing in autism, specially the finding that people with high functioning autism have an atypical cortical processing of voices when compared to typically developing kids. Based on these findings, the authors wanted to explore if theses differences in sound processing result in impaired ability to recognize the gender of a speaker base solely on voice perception.
The study included 20 adolescents with high functioning autism and 20 typically developing adolescents who were matched for IQ, gender, and age. Diagnoses were based on DSM-IV criteria and confirmed via ADI. The participants were presented voice fragments via headphones and were instructed to determine whether the voice was of a male or a female. The authors used a specialized software to morph the gender of the voices and thus present fragments that slowly changed from stereotypically masculine to stereotypically feminine.
The authors did not find any group differences in the accuracy for identifying the correct gender. Typically developing kids and those with high functioning autism were as accurate when identifying the gender of voice fragments. However, significant differences were observed in the speed of responding but only for the originally male morphed voices (male to female morph).
Please note however, that the differences in response time were not simple a 'mean' difference with those with HFA responding slower than typically developing kids. Instead, typically developing kids had a linear increase in response time as the voices were morphed. For example, they responded much faster when the voice was not morphed (original male voice presented as male voice) but gradually increased their response time as the voices were morphed, so that the slowest response time was when the voice was completely morphed (original male voice presented as a female voice). However, for those with HFA the effect was not linear. They responded fast when the voice was not morphed, slowed as the voice began to get morphed, but then became fast again when the voice was completely morphed.
The authors conclude that children with high functioning autism are not impaired in the perception of voice gender, but instead use a different perceptual approach resulting in different response times.